Discover AI Courses

Code Chain of Thought with Self-Evolution LLM: rStar-Math Mathematical Reasoning Framework

Delve into Microsoft's rStar-Math framework, exploring how small language models can achieve advanced mathematical reasoning through self-evolution and Monte Carlo Tree Search techniques.

• YouTube
• 34 minutes
• Self-Paced
• Free Video

Test-Time Preference Optimization: On-the-Fly AI Alignment via Iterative Feedback

Explore how Test-Time Preference Optimization enables AI models to dynamically align with human values through iterative feedback, without requiring traditional retraining methods.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

Building Complex Multi-Agent Configurations with Anthropic, Pydantic, and TogetherAI

Explore practical implementations of multi-agent AI systems using Anthropic's cookbook, PydanticAI framework, and Together.AI's demos for production-grade applications and real-world scenarios.

• YouTube
• 17 minutes
• Self-Paced
• Free Video

DeepSeek R1-Distill-Qwen Language Models - A Technical Overview and Comparison

Explore the groundbreaking DeepSeek R1 family of open-source reasoning models, comparing performance benchmarks with Sonnet 3.5, OpenAI o1, and other leading language models.

• YouTube
• 16 minutes
• Self-Paced
• Free Video

Intelligent Chain-of-Thought Router in Mixture of Experts with GraphLoRA - Technical Overview

Explore cutting-edge developments in MoE architecture, focusing on GraphLoRA integration, GNN routers, and recent advancements in PEFT implementations for enhanced AI model performance.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

Adaptive Reasoning AI: AdaptThink and ThinkLess Methods

Discover two groundbreaking AI reasoning methods: AdaptThink and ThinkLess, which teach language models to determine when complex reasoning is needed versus when simpler approaches suffice.

• YouTube
• 31 minutes
• Self-Paced
• Free Video

From DSPy to the CoT Encyclopedia: Improving AI Reasoning Capabilities

Explore how to enhance AI reasoning capabilities using DSPy, MCP, A2A, and other strategies through the CoT Encyclopedia, which offers insights on predicting and controlling reasoning behaviors in AI systems.

• YouTube
• 35 minutes
• Self-Paced
• Free Video

1984 for AI: Real-Time Thought Correction in AI Agents

Discover how real-time thought correction enhances AI agent safety, exploring techniques that allow AI systems to rethink and correct potentially harmful actions before execution.

• YouTube
• 17 minutes
• Self-Paced
• Free Video

Surprising Performance of Small Qwen3-A3B MoE - Testing on Extreme Logic Test

Explore the performance of the small Qwen3-A3B MoE model on an extreme logic test, comparing results with and without "thinking mode" in this live demonstration.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

Stabilizing Reasoning in Medical LLMs with Continued Pretraining and Reasoning Preference Optimization

Discover how to enhance medical LLMs through continued pre-training with domain knowledge in Japanese and preference optimization for stable reasoning, using a 72B Qwen2.5 model as the foundation.

• YouTube
• 25 minutes
• Self-Paced
• Free Video

Optimizing LLM Action Planning Space with In-Context Learning

Discover how In-Context Learning (ICL) can optimize action planning for LLMs, enhancing AI systems' ability to determine effective action sequences for reaching objectives in virtual and real environments.

• YouTube
• 15 minutes
• Self-Paced
• Free Video

Uncertainty AI: Breakthrough for Human-AI Reasoning

Discover how Google's Uncertainty AI framework in MedAI creates a blueprint for human-aligned AI by implementing iterative, transparent reasoning processes that mirror clinical decision-making in high-stakes domains.

• YouTube
• 14 minutes
• Self-Paced
• Free Video

A Smarter Way to Fine-Tune LLMs - Unlock Deeper Reasoning

Discover a smarter approach to fine-tuning Large Language Models that enhances their reasoning capabilities and generalization from in-context learning, based on research from Google DeepMind.

• YouTube
• 19 minutes
• Self-Paced
• Free Video

Re-Coding Reality: The Future of Vision Language Agents

Discover how theorem provers and digital twin representations are shaping next-generation AI systems in this Johns Hopkins research presentation on vision language agents.

• YouTube
• 32 minutes
• Self-Paced
• Free Video

HyperGRAPHS: Exploding Node-Dimensions, Hyperedges for Advanced AI Planning

Explore advanced AI planning techniques with HyperTrees, moving beyond Chain-of-Thoughts and Tree-of-Thoughts to enhance LLM performance for complex reasoning tasks.

• YouTube
• 23 minutes
• Self-Paced
• Free Video